paper-with-me

홈 › Papers

Catch Missing Details: Image Reconstruction with Frequency Augmented Variational Autoencoder

2023-05-04 · CVPR 2023 1 · Xinmiao Lin, Yikang Li, Jenhao Hsiao, Chiuman Ho, Yu Kong

The popular VQ-VAE models reconstruct images through learning a discrete codebook but suffer from a significant issue in the rapid quality degradation of image reconstruction as the compression rate rises. One major reason is that a higher compression rate induces more loss of visual signals on the higher frequency spectrum which reflect the details on pixel space. In this paper, a Frequency Complement Module (FCM) architecture is proposed to capture the missing frequency information for enhancing reconstruction quality. The FCM can be easily incorporated into the VQ-VAE structure, and we refer to the new model as Frequency Augmented VAE (FA-VAE). In addition, a Dynamic Spectrum Loss (DSL) is introduced to guide the FCMs to balance between various frequencies dynamically for optimal reconstruction. FA-VAE is further extended to the text-to-image synthesis task, and a Cross-attention Autoregressive Transformer (CAT) is proposed to obtain more precise semantic attributes in texts. Extensive reconstruction experiments with different compression rates are conducted on several benchmark datasets, and the results demonstrate that the proposed FA-VAE is able to restore more faithfully the details compared to SOTA methods. CAT also shows improved generation quality with better image-text semantic alignment.

📄 PDF Abstract BibTeX arXiv:2305.02541

Code (1)

oppo-us-research/FA-VAE 공식 구현 pytorch

Tasks

Image GenerationImage Reconstruction

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Adam 설명 없음
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Multi-Head Attention 설명 없음

Similar Papers 제목 키워드 기반

3D Surface Reconstruction with Enhanced High-Frequency Details

2025-05-06 · Shikun Zhang, Yiqun Wang, Cunjian Chen, Yong Li 외

Neural implicit 3D reconstruction can reproduce shapes without 3D supervision, and it learns the 3D scene through volume rendering methods and neural implicit representations. Current neural surface reconstruction method…

3D ReconstructionSurface Reconstruction

CATCH: Channel-Aware multivariate Time Series Anomaly Detection via Frequency Patching

2024-10-16 · Xingjian Wu, Xiangfei Qiu, Zhengyu Li, Yihang Wang 외

Anomaly detection in multivariate time series is challenging as heterogeneous subsequence anomalies may occur. Reconstruction-based methods, which focus on learning normal patterns in the frequency domain to detect diver…

Anomaly DetectionTime SeriesTime Series Anomaly Detection

FreeSeed: Frequency-band-aware and Self-guided Network for Sparse-view CT Reconstruction

2023-07-12 · Chenglong Ma, Zilong Li, Junping Zhang, Yi Zhang 외

Sparse-view computed tomography (CT) is a promising solution for expediting the scanning process and mitigating radiation exposure to patients, the reconstructed images, however, contain severe streak artifacts, compromi…

Computed Tomography (CT)CT Reconstruction

Frequency-Aware Reconstruction of Fluid Simulations with Generative Networks

2019-12-18 · Simon Biland, Vinicius C. Azevedo, Byung-soo Kim, Barbara Solenthaler

Convolutional neural networks were recently employed to fully reconstruct fluid simulation data from a set of reduced parameters. However, since (de-)convolutions traditionally trained with supervised L1-loss functions d…

DiffSR: Learning Radar Reflectivity Synthesis via Diffusion Model from Satellite Observations

2024-11-11 · Xuming He, Zhiwang Zhou, Wenlong Zhang, Xiangyu Zhao 외

Weather radar data synthesis can fill in data for areas where ground observations are missing. Existing methods often employ reconstruction-based approaches with MSE loss to reconstruct radar data from satellite observat…